> We have an application made up of 15 processes that interact through
> shared memories and sockets. Each process computes for a few
> milliseconds and then pass its data to the next one (a pipeline
> computation scheme). When a process is done with its computation, it
> calls pause() to release the CPU, it resumes its computation when its
> timer interrupt wakes it every 10 milliseconds, if the computation
> were not completed, we log this as an overrun.
could you try the attached patch? It's a fixed/cleaned up version of Paul
Campbell <paul@chromatic.com>'s 2.2.0 patch, with this patch we
immediately send signal events to other CPUs too. It's against recent
2.2.[01] kernels.
-- mingo
--- linux/kernel/signal.c.orig Tue Dec 29 16:37:02 1998
+++ linux/kernel/signal.c Fri Feb 19 13:14:20 1999
@@ -363,8 +363,27 @@
}
sigaddset(&t->signal, sig);
- if (!sigismember(&t->blocked, sig))
+ if (!sigismember(&t->blocked, sig)) {
t->sigpending = 1;
+#ifdef __SMP__
+ /*
+ * If the task is running on a different CPU
+ * force a reschedule on the other CPU - note that
+ * the code below is a tad loose and might occasionally
+ * kick the wrong CPU if we catch the process in the
+ * process of changing - but no harm is done by that
+ * other than doing an extra (lightweight) IPI interrupt.
+ *
+ * note that we rely on the previous spin_lock to
+ * lock interrupts for us! No need to set need_resched
+ * since signal event passing goes through ->blocked.
+ */
+ spin_lock(&runqueue_lock);
+ if (t->has_cpu && t->processor != smp_processor_id())
+ smp_send_reschedule(t->processor);
+ spin_unlock(&runqueue_lock);
+#endif /* __SMP__ */
+ }
out:
spin_unlock_irqrestore(&t->sigmask_lock, flags);
--- linux/include/linux/sched.h.orig Fri Feb 19 13:07:51 1999
+++ linux/include/linux/sched.h Fri Feb 19 13:08:38 1999
@@ -112,6 +112,7 @@
* a separate lock).
*/
extern rwlock_t tasklist_lock;
+extern spinlock_t runqueue_lock;
extern spinlock_t scheduler_lock;
extern void sched_init(void);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/